Four Questions for  ECE's Zhao Zhang

Before joining the School of Engineering Department of Electrical and Computer Engineering as an assistant professor in fall 2023, Zhao Zhang led the machine learning group at Texas Advanced Computing Center (TACC). After receiving his PhD in computer science from the University of Chicago and before his leadership role at TACC, he was a postdoc researcher at UC Berkeley’s AMPLab, and a data science fellow at the Berkeley Institute for Data Science. He applies his extensive experience in high-performance computing (HPC) and big data systems to his recent research focus on the fusion of HPC and deep learning (DL) with a wide range of topics of optimization algorithm, I/O, architecture, and domain applications. 

Four Questions for Assistant Professor Zhao Zhang 

What fuels your passion for electrical and computer engineering? 
I have long been interested in high-performance computing algorithms and systems, and their applications in science and engineering. Over my career, I have been working on the world’s most powerful supercomputers, including Blue Gene/P, Stampede 2/3, Frontera, Polaris, and Perlmutter. I have also been collaborating with domain scientists from astronomy and  neural science to structural biology.  

What is your current research focus -- and who is most likely to benefit from its results? 
My current research focuses on the optimization algorithms and systems for large-scale neural network pretraining on thousands to tens of thousands of graphics processing units, or GPUs. I am also active in next-generation national AI infrastructure design.  

Deep learning practitioners from academia and industry are likely to benefit from my research with optimized training turnaround, resource usage, and ease of neural network training and inference. National computing centers will be able to leverage my research results to better support users while also enhancing overall machine utilization. 
 

What have you most enjoyed about the ECE department and the School of Engineering? 
I enjoy the PODS project led by Wade Trappe, which encourages interdisciplinary research among young faculty members. All junior faculty members like me, Rutgers Artificial Intelligence and Data Science Collaboratory who joined the school in 2023, are part of it. We used to meet every few weeks during the 2023-24 academic year. I have been collaborating with my colleagues on a few projects and funding opportunities, including working with Ashley Guo, an assistant professor of chemical and biochemical engineering on an NSF project, and providing computing resources for Rutgers assistant professor of statistics, Matteo Bonvini. 

What has surprised you the most about the school and your department?  
I am surprised by the vision and effort directed by ECE and SOE towards interdisciplinary research, as it is not a common research goal in other universities. With the recently announced Rutgers Artificial Intelligence and Data Science Collaboratory – a university-wide hub for data science, AI, and student and community engagement – we’ll able to perform next-level, highly integrated interdisciplinary research across many areas.